Yulia Belov

The world runs on sequences. Every meaningful interaction - in games, with agents, in robotics - unfolds as ordered events over time. Most current systems were built for something simpler. That mismatch is starting to limit what’s actually possible. #SequentialReality

A nuclear plant only gets built where there’s massive, constant demand. Nuclear fission was the revolution. The plant was not. AI is the same. The ability to compute LLMs is the revolution. But few players burning capital to keep the lights on is not.

A real revolution changes everything. Java VM appeared, apps multiplied, dependence on the underlying HW vanished. Unity appeared, gaming exploded, nobody had to build their own engine anymore. AI? A few winners, no wave of startups building their own models. No revolution yet.

I keep hearing “development costs are zero now.” Wrong. That’s said by people who’ve never shipped production quality. Code that demos is free. Code that survives a million users in production is not.

@nifinet team

True. Internal models win. A company owns its IP, and on that IP plus its human capital, it builds its own models. This requires the right infrastructure. But the payoff is huge: you keep your knowledge in-house and stop depending on the model providers.

@rexsalisbury Home reno is the only project where “almost done” is a permanent state of being.

my biggest value as a VC is this clause in my standard TS. iykyk.

Here’s what nobody admits. The problem existing was never the hard part. The hard part is who actually needs it and whether you can ship it. LLMs? Technical achievement. Agents? Scaling economics. This isn’t a deep tech challenge anymore. It’s distribution and judgment.

Linus is right. AI is just a tool, like compilers. AI gives individuals enormous autonomy. But the most important thing to transfer was never the code. It’s the principles of how people work together. Everything we do is still people to people.

@adamshuaib Exactly. Classic corporate conflict of interests. When you’re smarter than your boss, you see risks they can’t. Flag them and you’re labeled a “no-man.” So smart people have two exits: switch companies or start their own.

The founders worth betting on were often terrible employees. Many clashed with previous managers, ignored instructions and quickly realised they were smarter than their boss. In the wrong environment these people are marginalised or ignored. It’s also why previous employer references for a founder don’t always show their true ability. This kind of behaviour doesn’t tend to show up well on a CV. It looks like job-hopping or not being a team player. But in practice it signals someone who is perfectly suited to the unstructured, uncertain nature of running a startup. Our data on 15,000 companies showed that founders who scored highly on emotional stability (not charisma or likability) were far more likely to build bigger companies.

@MartinGTobias Exactly. The words you use define how you see the world. “I choose to” makes you the actor. “I have to” makes you the audience. That’s not grammar. That’s agency.

I don't "have to" respond to emails. I get to. Language creates reality. "Have to" is victim language. Choose your words.

@SahilBloom Reminds of Vadim Zeland and his Reality Transurfing. Literally same idea came to different people in different societies. I would also add the lucky experiment - when people are open to the new, they see more opportunities.

Everyone needs to read this... Black Coffee Theory (a visual thread)

Companies overpay for features they never use. Cloud, sales stack, database. All oversized. The winners aren’t the most powerful tools. They’re the ones customers actually use. Notion proved it. Ship what people need. Nothing more. Let them pay for value, not features.

Companies burn millions on solutions that don’t fit. AWS stopped making sense for most startups long ago. Cloudflare, DigitalOcean, Hetzner are simpler and cheaper for what most actually need. Same for databases and storage. The defaults are expensive. The alternatives are right there.

@StephNass Nobody funds a deck. A deck is just an invitation to a conversation.

Founder: "Our deck is the problem." Reality: Your traction is the problem.

@Jbm_dev The market has a missing layer. We see it. The future demands it. We know how to build it.

why are you building what your building? not the polished answer. the real one.

Everyone ships what they’re capable of, not what the market demands. From the outside it reads as genius or incompetence. It’s neither. It’s capability ceiling. Same reason most hairdressers give you their cut, not your cut. The skill gap is the product gap.

@0xJeff I ran product for two L1 chains. Both had great vision. One founder chased hype and lost focus on what mattered. The other shipped part of the vision but underestimated external risks. Both were dead within a year of public launch.

The longer I invest in crypto, the more I realize this industry is mostly a founder filtering machine ​ (a lot more challenging than in Trad VC/PE) ​ After 50+ investments + years of advising founders, I'm starting to see the patterns. The patterns that separate founders that bring return vs founders that don't. ​ In today's piece, we'll go over - My actual investment theses/outcomes from 2025–2026 - What worked, what completely failed, and why - The founder traits that correlate with return ​ This is probably the most honest breakdown I’ve written so far. May add value to you as an investor or builder navigating the space in 2026. ​ NFA/DYOR ​ [Link in Bio]

@saltzman_jason @OpenAI @RonnieChatterji This is the data point that matters. Software ate the world, and now AI is being eaten by the world right back. 95% of AI entrepreneurship is outside tech.

The businesses getting the most out of AI right now aren't tech companies. They’re plumbers, agency owners, dentists, Etsy sellers. New data from @OpenAI (cc @RonnieChatterji) makes the pattern clear: tech startups account for about 5% of ~active~ U.S. users doing entrepreneurial work with ChatGPT. The other 95% are spread across services, retail, healthcare, and trades. AI adoption for entrepreneurship isn’t concentrated in tech. It’s happening inside everyday businesses, folding into routine work that used to be slower or outsourced or entirely overlooked.